{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:HIQIFA5URNEMIYIFV5YCG3EA3O","short_pith_number":"pith:HIQIFA5U","canonical_record":{"source":{"id":"2407.00066","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-17T15:21:35Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"324a325cbd21422fe1880be6e0b01c103ba28b0f443ec7b79696d35272d50c31","abstract_canon_sha256":"a4e7f7053b62d6409988cddb591e70854f9eacf941b219bf1f55b95887869a06"},"schema_version":"1.0"},"canonical_sha256":"3a208283b48b48c46105af70236c80db92c3814a33f852d725de204f8bb21513","source":{"kind":"arxiv","id":"2407.00066","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.00066","created_at":"2026-07-05T11:12:18Z"},{"alias_kind":"arxiv_version","alias_value":"2407.00066v4","created_at":"2026-07-05T11:12:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.00066","created_at":"2026-07-05T11:12:18Z"},{"alias_kind":"pith_short_12","alias_value":"HIQIFA5URNEM","created_at":"2026-07-05T11:12:18Z"},{"alias_kind":"pith_short_16","alias_value":"HIQIFA5URNEMIYIF","created_at":"2026-07-05T11:12:18Z"},{"alias_kind":"pith_short_8","alias_value":"HIQIFA5U","created_at":"2026-07-05T11:12:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:HIQIFA5URNEMIYIFV5YCG3EA3O","target":"record","payload":{"canonical_record":{"source":{"id":"2407.00066","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-17T15:21:35Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"324a325cbd21422fe1880be6e0b01c103ba28b0f443ec7b79696d35272d50c31","abstract_canon_sha256":"a4e7f7053b62d6409988cddb591e70854f9eacf941b219bf1f55b95887869a06"},"schema_version":"1.0"},"canonical_sha256":"3a208283b48b48c46105af70236c80db92c3814a33f852d725de204f8bb21513","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:18.425342Z","signature_b64":"YrK0A0YPSrMFn1y2rkNprWWq7FzX920AEAgWKrWlzuns3SDM7ic00ErmcASkGjlrKzH9QCwZotfEoMEOOxyOBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a208283b48b48c46105af70236c80db92c3814a33f852d725de204f8bb21513","last_reissued_at":"2026-07-05T11:12:18.424774Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:18.424774Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.00066","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:12:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6nTCv3TJLlYHYqeGJS/pspJ5SbdfqwKAW1oc7TVvj74nQ6ArfV3kVl98ubIJUDEtHwBp53OMsZ90EtFn3gtiDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T01:33:06.236869Z"},"content_sha256":"67fc2744debfd5d6eee35d7268c2778852d8f5c018b85cf7260cd5a7cede0d89","schema_version":"1.0","event_id":"sha256:67fc2744debfd5d6eee35d7268c2778852d8f5c018b85cf7260cd5a7cede0d89"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:HIQIFA5URNEMIYIFV5YCG3EA3O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.DC","authors_text":"Jiacheng Zhu, Justin Solomon, Kristjan Greenewald, Leshem Choshen, Mikhail Yurochkin, Onkar Bhardwaj, Rickard Br\\\"uel-Gabrielsson","submitted_at":"2024-06-17T15:21:35Z","abstract_excerpt":"Fine-tuning large language models (LLMs) with low-rank adaptations (LoRAs) has become common practice, often yielding numerous copies of the same LLM differing only in their LoRA updates. This paradigm presents challenges for systems that serve real-time responses to queries that each involve a different LoRA. Prior works optimize the design of such systems but still require continuous loading and offloading of LoRAs, as it is infeasible to store thousands of LoRAs in GPU memory. To mitigate this issue, we investigate the efficacy of compression when serving LoRAs. We propose a method for the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.00066","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2407.00066/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:12:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EpYuYybzeIv6pm1cN6VyydeBOqAwaUt6WvKXUQqUIcHHaS62CW+inP0/QmvnocbrEsliqs9hz4ZZ0dsmKRmGBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T01:33:06.237195Z"},"content_sha256":"be076bd1ea433a5db58252e771e9f29c4650da0bd814b1df4de53f5aa07aa92f","schema_version":"1.0","event_id":"sha256:be076bd1ea433a5db58252e771e9f29c4650da0bd814b1df4de53f5aa07aa92f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HIQIFA5URNEMIYIFV5YCG3EA3O/bundle.json","state_url":"https://pith.science/pith/HIQIFA5URNEMIYIFV5YCG3EA3O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HIQIFA5URNEMIYIFV5YCG3EA3O/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-15T01:33:06Z","links":{"resolver":"https://pith.science/pith/HIQIFA5URNEMIYIFV5YCG3EA3O","bundle":"https://pith.science/pith/HIQIFA5URNEMIYIFV5YCG3EA3O/bundle.json","state":"https://pith.science/pith/HIQIFA5URNEMIYIFV5YCG3EA3O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HIQIFA5URNEMIYIFV5YCG3EA3O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:HIQIFA5URNEMIYIFV5YCG3EA3O","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"a4e7f7053b62d6409988cddb591e70854f9eacf941b219bf1f55b95887869a06","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-17T15:21:35Z","title_canon_sha256":"324a325cbd21422fe1880be6e0b01c103ba28b0f443ec7b79696d35272d50c31"},"schema_version":"1.0","source":{"id":"2407.00066","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.00066","created_at":"2026-07-05T11:12:18Z"},{"alias_kind":"arxiv_version","alias_value":"2407.00066v4","created_at":"2026-07-05T11:12:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.00066","created_at":"2026-07-05T11:12:18Z"},{"alias_kind":"pith_short_12","alias_value":"HIQIFA5URNEM","created_at":"2026-07-05T11:12:18Z"},{"alias_kind":"pith_short_16","alias_value":"HIQIFA5URNEMIYIF","created_at":"2026-07-05T11:12:18Z"},{"alias_kind":"pith_short_8","alias_value":"HIQIFA5U","created_at":"2026-07-05T11:12:18Z"}],"graph_snapshots":[{"event_id":"sha256:be076bd1ea433a5db58252e771e9f29c4650da0bd814b1df4de53f5aa07aa92f","target":"graph","created_at":"2026-07-05T11:12:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2407.00066/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuning large language models (LLMs) with low-rank adaptations (LoRAs) has become common practice, often yielding numerous copies of the same LLM differing only in their LoRA updates. This paradigm presents challenges for systems that serve real-time responses to queries that each involve a different LoRA. Prior works optimize the design of such systems but still require continuous loading and offloading of LoRAs, as it is infeasible to store thousands of LoRAs in GPU memory. To mitigate this issue, we investigate the efficacy of compression when serving LoRAs. We propose a method for the ","authors_text":"Jiacheng Zhu, Justin Solomon, Kristjan Greenewald, Leshem Choshen, Mikhail Yurochkin, Onkar Bhardwaj, Rickard Br\\\"uel-Gabrielsson","cross_cats":["cs.AI","cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-17T15:21:35Z","title":"Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.00066","kind":"arxiv","version":4},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:67fc2744debfd5d6eee35d7268c2778852d8f5c018b85cf7260cd5a7cede0d89","target":"record","created_at":"2026-07-05T11:12:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"a4e7f7053b62d6409988cddb591e70854f9eacf941b219bf1f55b95887869a06","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-17T15:21:35Z","title_canon_sha256":"324a325cbd21422fe1880be6e0b01c103ba28b0f443ec7b79696d35272d50c31"},"schema_version":"1.0","source":{"id":"2407.00066","kind":"arxiv","version":4}},"canonical_sha256":"3a208283b48b48c46105af70236c80db92c3814a33f852d725de204f8bb21513","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3a208283b48b48c46105af70236c80db92c3814a33f852d725de204f8bb21513","first_computed_at":"2026-07-05T11:12:18.424774Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:12:18.424774Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YrK0A0YPSrMFn1y2rkNprWWq7FzX920AEAgWKrWlzuns3SDM7ic00ErmcASkGjlrKzH9QCwZotfEoMEOOxyOBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:12:18.425342Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.00066","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:67fc2744debfd5d6eee35d7268c2778852d8f5c018b85cf7260cd5a7cede0d89","sha256:be076bd1ea433a5db58252e771e9f29c4650da0bd814b1df4de53f5aa07aa92f"],"state_sha256":"7dccca58768d570d725aa5cf9f43abb9022821d549cc655616ba500dc56d7365"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XnPDRsKP3vueqWVdsX/RRA/0SD+bgTPUFwiXXtHMW7PVN9XFmiZcqPfyLqNTFCgCbqsVmu8jKekq2tQqzp3RCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T01:33:06.240899Z","bundle_sha256":"196bc50f66d6b898026f8cfb7d281a666e90bdbc53b68dcd3be58137e071f7eb"}}